feat: add csv upload functionality for OSS (#29)

feat: add csv upload functionality
chore: remove redundant arq-worker from docker-compose

.dockerignore

@@ -0,0 +1 @@
+api/.env

api/Dockerfile

@@ -11,7 +11,7 @@ RUN apt-get update && apt-get install -y \
     && rm -rf /var/lib/apt/lists/*
 
 # Copy and install requirements
-COPY requirements.txt .
+COPY api/requirements.txt .
 
 # Install dependencies to user directory for easy copying
 RUN pip install --user --no-cache-dir -r requirements.txt && \
@@ -53,7 +53,8 @@ ENV PYTHONDONTWRITEBYTECODE=1
 ENV PYTHONUNBUFFERED=1
 
 # Copy application code
-COPY . ./api
+COPY ./api ./api
+COPY ./scripts/start_services.sh ./scripts/start_services.sh
 
 ENV PYTHONPATH=/app
 
@@ -61,4 +62,4 @@ ENV PYTHONPATH=/app
 EXPOSE 8000
 
 # Run the FastAPI app with uvicorn
-CMD ["uvicorn", "api.app:app", "--host", "0.0.0.0", "--port", "8000"]
+CMD ["bash", "-c", "./scripts/start_services.sh && tail -f ./logs/latest/*.log"]

api/app.py

@@ -2,29 +2,22 @@
 
 import sentry_sdk
 
-from api.constants import ENABLE_SENTRY, REDIS_URL, SENTRY_DSN
+from api.constants import DEPLOYMENT_MODE, ENABLE_TELEMETRY, REDIS_URL, SENTRY_DSN
 from api.logging_config import ENVIRONMENT, setup_logging
 
 # Set up logging and get the listener for cleanup
-logging_queue_listener = setup_logging()
+setup_logging()
 
 
-if ENABLE_SENTRY:
-    if not SENTRY_DSN:
-        print(
-            "Warning: ENABLE_SENTRY is true but SENTRY_DSN is not configured. Sentry disabled."
-        )
-    else:
-        sentry_sdk.init(
-            dsn=SENTRY_DSN,
-            # Add data like request headers and IP for users,
-            # see https://docs.sentry.io/platforms/python/data-management/data-collected/ for more info
-            send_default_pii=True,
-            environment=ENVIRONMENT,
-        )
-        print(f"Sentry initialized in environment: {ENVIRONMENT}")
-else:
-    print(f"Sentry disabled (ENABLE_SENTRY=false)")
+if SENTRY_DSN and (
+    DEPLOYMENT_MODE != "oss" or (DEPLOYMENT_MODE == "oss" and ENABLE_TELEMETRY)
+):
+    sentry_sdk.init(
+        dsn=SENTRY_DSN,
+        send_default_pii=True,
+        environment=ENVIRONMENT,
+    )
+    print(f"Sentry initialized in environment: {ENVIRONMENT}")
 
 
 import asyncio
@@ -91,9 +84,6 @@ async def lifespan(app: FastAPI):
 
     await redis.aclose()
 
-    if logging_queue_listener is not None:
-        logging_queue_listener.stop()
-
 
 app = FastAPI(
     title="Dograh API",

api/constants.py

@@ -43,5 +43,9 @@ S3_BUCKET = os.environ.get("S3_BUCKET")
 S3_REGION = os.environ.get("S3_REGION", "us-east-1")
 
 # Sentry configuration
-ENABLE_SENTRY = os.getenv("ENABLE_SENTRY", "false").lower() == "true"
 SENTRY_DSN = os.getenv("SENTRY_DSN")
+
+
+ENABLE_ARI_STASIS = os.getenv("ENABLE_ARI_STASIS", "false").lower() == "true"
+SERIALIZE_LOG_OUTPUT = os.getenv("SERIALIZE_LOG_OUTPUT", "false").lower() == "true"
+ENABLE_TELEMETRY = os.getenv("ENABLE_TELEMETRY", "false").lower() == "true"

api/logging_config.py

@@ -3,6 +3,7 @@ import sys
 
 import loguru
 
+from api.constants import SERIALIZE_LOG_OUTPUT
 from api.enums import Environment
 from api.utils.worker import get_worker_id, is_worker_process
 from pipecat.utils.context import run_id_var, turn_var
@@ -88,7 +89,7 @@ def setup_logging():
         patched.add(
             actual_log_path,
             level=log_level,
-            serialize=True,  # Use JSON serialization for structured logs
+            serialize=SERIALIZE_LOG_OUTPUT,  # Use JSON serialization for structured logs
             enqueue=True,  # Thread-safe writing
             backtrace=True,  # Include full traceback in exceptions
             diagnose=False,  # Don't include local variables in traceback for security

api/routes/campaign.py

@@ -9,6 +9,7 @@ from api.db.models import UserModel
 from api.enums import OrganizationConfigurationKey
 from api.services.auth.depends import get_user
 from api.services.campaign.runner import campaign_runner_service
+from api.services.storage import storage_fs
 
 router = APIRouter(prefix="/campaign")
 
@@ -16,7 +17,8 @@ router = APIRouter(prefix="/campaign")
 class CreateCampaignRequest(BaseModel):
     name: str = Field(..., min_length=1, max_length=255)
     workflow_id: int
-    source_id: str  # Sheet URL
+    source_type: str = Field(..., pattern="^(google-sheet|csv)$")
+    source_id: str  # Google Sheet URL or CSV file key
 
 
 class CampaignResponse(BaseModel):
@@ -74,7 +76,7 @@ async def create_campaign(
     campaign = await db_client.create_campaign(
         name=request.name,
         workflow_id=request.workflow_id,
-        source_type="google-sheet",
+        source_type=request.source_type,
         source_id=request.source_id,
         user_id=user.id,
         organization_id=user.selected_organization_id,
@@ -174,14 +176,10 @@ async def start_campaign(
         OrganizationConfigurationKey.TWILIO_CONFIGURATION.value,
     )
 
-    if (
-        not twilio_config
-        or not twilio_config.value
-        or not twilio_config.value.get("value")
-    ):
+    if not twilio_config or not twilio_config.value:
         raise HTTPException(
             status_code=401,
-            detail="Your organisation is not allowed to make phone call. Contact founders@dograh.com for further support.",
+            detail="You must configure telephony first by going to APP_URL/configure-telephony",
         )
 
     # Verify campaign exists and belongs to organization
@@ -286,14 +284,10 @@ async def resume_campaign(
         OrganizationConfigurationKey.TWILIO_CONFIGURATION.value,
     )
 
-    if (
-        not twilio_config
-        or not twilio_config.value
-        or not twilio_config.value.get("value")
-    ):
+    if not twilio_config or not twilio_config.value:
         raise HTTPException(
             status_code=401,
-            detail="Your organisation is not allowed to make phone call. Contact founders@dograh.com for further support.",
+            detail="You must configure telephony first by going to APP_URL/configure-telephony",
         )
 
     # Verify campaign exists and belongs to organization
@@ -345,3 +339,59 @@ async def get_campaign_progress(
         return CampaignProgressResponse(**progress)
     except ValueError as e:
         raise HTTPException(status_code=400, detail=str(e))
+
+
+class CampaignSourceDownloadResponse(BaseModel):
+    download_url: str
+    expires_in: int
+
+
+@router.get("/{campaign_id}/source-download-url")
+async def get_campaign_source_download_url(
+    campaign_id: int,
+    user: UserModel = Depends(get_user),
+) -> CampaignSourceDownloadResponse:
+    """Get presigned download URL for campaign CSV source file
+
+    Only works for CSV source type. For Google Sheets, use the source_id directly.
+    Validates that the campaign belongs to the user's organization for security.
+    """
+    # Verify campaign exists and belongs to organization
+    campaign = await db_client.get_campaign(campaign_id, user.selected_organization_id)
+    if not campaign:
+        raise HTTPException(status_code=404, detail="Campaign not found")
+
+    # Only generate download URL for CSV files
+    if campaign.source_type != "csv":
+        raise HTTPException(
+            status_code=400,
+            detail=f"Download URL only available for CSV sources. This campaign uses {campaign.source_type}",
+        )
+
+    # Verify the file key belongs to the user's organization
+    # File key format: campaigns/{org_id}/{uuid}_{filename}.csv
+    if not campaign.source_id.startswith(f"campaigns/{user.selected_organization_id}/"):
+        raise HTTPException(
+            status_code=403,
+            detail="Access denied: Source file does not belong to your organization",
+        )
+
+    # Generate presigned download URL
+    try:
+        download_url = await storage_fs.aget_signed_url(
+            campaign.source_id,
+            expiration=3600,  # 1 hour
+        )
+
+        if not download_url:
+            raise HTTPException(
+                status_code=500, detail="Failed to generate download URL"
+            )
+
+        return CampaignSourceDownloadResponse(
+            download_url=download_url, expires_in=3600
+        )
+    except Exception as e:
+        raise HTTPException(
+            status_code=500, detail=f"Failed to generate download URL: {str(e)}"
+        )

api/routes/s3_signed_url.py

@@ -1,8 +1,11 @@
+import re
+import uuid
 from typing import Annotated, Any, Dict, Optional, TypedDict
 
 from botocore.exceptions import ClientError
 from fastapi import APIRouter, Depends, HTTPException, Query
 from loguru import logger
+from pydantic import BaseModel, Field
 
 from api.db import db_client
 from api.enums import StorageBackend
@@ -20,6 +23,20 @@ class FileMetadataResponse(TypedDict):
     metadata: Optional[Dict[str, Any]]
 
 
+class PresignedUploadUrlRequest(BaseModel):
+    file_name: str = Field(..., pattern=r".*\.csv$", description="CSV filename")
+    file_size: int = Field(
+        ..., gt=0, le=10_485_760, description="File size in bytes (max 10MB)"
+    )
+    content_type: str = Field(default="text/csv", description="File content type")
+
+
+class PresignedUploadUrlResponse(BaseModel):
+    upload_url: str
+    file_key: str
+    expires_in: int
+
+
 router = APIRouter(prefix="/s3", tags=["s3"])
 
 
@@ -217,3 +234,65 @@ async def get_file_metadata(
     except Exception as exc:
         logger.error(f"Error getting file metadata: {exc}")
         raise HTTPException(status_code=500, detail="Failed to get file metadata")
+
+
+@router.post(
+    "/presigned-upload-url",
+    response_model=PresignedUploadUrlResponse,
+    summary="Generate a presigned URL for direct CSV upload",
+)
+async def get_presigned_upload_url(
+    request: PresignedUploadUrlRequest,
+    user=Depends(get_user),
+):
+    """Generate a presigned PUT URL for direct CSV file upload to S3/MinIO.
+
+    This endpoint enables browser-to-storage uploads without passing through the backend
+
+    Access Control:
+    * All authenticated users can upload CSV files scoped to their organization.
+    * Files are stored with organization-scoped keys for multi-tenancy.
+
+    Returns:
+    * upload_url: Presigned URL (valid for 15 minutes) for PUT request
+    * file_key: Unique storage key to use as source_id in campaign creation
+    * expires_in: URL expiration time in seconds
+    """
+
+    # Sanitize filename - remove special chars, keep only alphanumeric, dash, underscore, and dot
+    sanitized_name = re.sub(r"[^a-zA-Z0-9._-]", "_", request.file_name)
+
+    # Generate unique file key: campaigns/{org_id}/{uuid}_{filename}.csv
+    file_key = (
+        f"campaigns/{user.selected_organization_id}/{uuid.uuid4()}_{sanitized_name}"
+    )
+
+    try:
+        # Generate presigned PUT URL using current storage backend
+        upload_url = await storage_fs.aget_presigned_put_url(
+            file_path=file_key,
+            expiration=900,  # 15 minutes
+            content_type=request.content_type,
+            max_size=request.file_size,
+        )
+
+        if not upload_url:
+            raise HTTPException(
+                status_code=500, detail="Failed to generate presigned upload URL"
+            )
+
+        logger.info(
+            f"Generated presigned upload URL for user {user.id}, org {user.selected_organization_id}, file_key: {file_key}"
+        )
+
+        return PresignedUploadUrlResponse(
+            upload_url=upload_url,
+            file_key=file_key,
+            expires_in=900,
+        )
+
+    except Exception as exc:
+        logger.error(f"Error generating presigned upload URL: {exc}")
+        raise HTTPException(
+            status_code=500, detail="Failed to generate presigned upload URL"
+        )

api/services/campaign/campaign_orchestrator.py

@@ -7,7 +7,7 @@ for final completion after 1 hour of inactivity and handles retry events.
 
 from api.logging_config import setup_logging
 
-logging_queue_listener = setup_logging()
+setup_logging()
 
 
 import asyncio
@@ -495,7 +495,7 @@ class CampaignOrchestrator:
         if self._pubsub:
             try:
                 await self._pubsub.unsubscribe(RedisChannel.CAMPAIGN_EVENTS.value)
-                await self._pubsub.close()
+                await self._pubsub.aclose()
             except Exception as e:
                 logger.error(f"Error closing pubsub: {e}")
 
@@ -538,16 +538,12 @@ async def main():
         if shutdown_task in done:
             logger.info("Shutdown signal received, stopping orchestrator...")
             orchestrator._running = False
-            # Wait for orchestrator to finish gracefully
+            # Cancel the orchestrator task immediately since it may be blocked
+            orchestrator_task.cancel()
             try:
-                await asyncio.wait_for(orchestrator_task, timeout=5.0)
-            except asyncio.TimeoutError:
-                logger.warning("Orchestrator shutdown timeout, cancelling...")
-                orchestrator_task.cancel()
-                try:
-                    await orchestrator_task
-                except asyncio.CancelledError:
-                    pass
+                await orchestrator_task
+            except asyncio.CancelledError:
+                logger.info("Orchestrator task cancelled successfully")
 
     except KeyboardInterrupt:
         logger.info("Keyboard interrupt received")

api/services/campaign/source_sync.py

@@ -31,19 +31,3 @@ class CampaignSourceSyncService(ABC):
             f"Getting credentials for org {organization_id}, source {source_type}"
         )
         return {}
-
-
-def get_sync_service(source_type: str) -> CampaignSourceSyncService:
-    """Returns appropriate sync service based on source type"""
-    from .sources.google_sheets import GoogleSheetsSyncService
-
-    services = {
-        "google-sheet": GoogleSheetsSyncService,
-        # Add more as needed: "hubspot": HubSpotSyncService,
-    }
-
-    service_class = services.get(source_type)
-    if not service_class:
-        raise ValueError(f"Unknown source type: {source_type}")
-
-    return service_class()

api/services/campaign/source_sync_factory.py

@@ -0,0 +1,19 @@
+from api.services.campaign.source_sync import CampaignSourceSyncService
+from api.services.campaign.sources.csv import CSVSyncService
+from api.services.campaign.sources.google_sheets import GoogleSheetsSyncService
+
+
+def get_sync_service(source_type: str) -> CampaignSourceSyncService:
+    """Returns appropriate sync service based on source type"""
+
+    services = {
+        "google-sheet": GoogleSheetsSyncService,
+        "csv": CSVSyncService,
+        # Add more as needed: "hubspot": HubSpotSyncService,
+    }
+
+    service_class = services.get(source_type)
+    if not service_class:
+        raise ValueError(f"Unknown source type: {source_type}")
+
+    return service_class()

api/services/campaign/sources/csv.py

@@ -0,0 +1,140 @@
+import csv
+import hashlib
+from io import StringIO
+from typing import Any, Dict, List
+
+import httpx
+from loguru import logger
+
+from api.db import db_client
+from api.services.campaign.source_sync import CampaignSourceSyncService
+from api.services.storage import storage_fs
+
+
+class CSVSyncService(CampaignSourceSyncService):
+    """Implementation for CSV file synchronization"""
+
+    async def sync_source_data(self, campaign_id: int) -> int:
+        """
+        Fetches data from CSV file in S3/MinIO and creates queued_runs
+        """
+        # Get campaign
+        campaign = await db_client.get_campaign_by_id(campaign_id)
+        if not campaign:
+            raise ValueError(f"Campaign {campaign_id} not found")
+
+        # 1. Get download URL using internal endpoint (for container-to-container access)
+        file_key = campaign.source_id
+        signed_url = await storage_fs.aget_signed_url(
+            file_key, expiration=3600, use_internal_endpoint=True
+        )
+
+        if not signed_url:
+            raise ValueError(f"Failed to generate download URL for file: {file_key}")
+
+        # 2. Download CSV file
+        async with httpx.AsyncClient() as client:
+            try:
+                response = await client.get(signed_url)
+                response.raise_for_status()
+                csv_content = response.text
+            except httpx.HTTPError as e:
+                logger.error(f"Failed to download CSV file: {e} for url: {signed_url}")
+                raise ValueError(f"Failed to download CSV file from storage: {str(e)}")
+
+        # 3. Parse CSV
+        csv_data = self._parse_csv(csv_content)
+
+        if not csv_data or len(csv_data) < 2:
+            logger.warning(f"No data found in CSV for campaign {campaign_id}")
+            return 0
+
+        headers = csv_data[0]  # First row is headers
+        rows = csv_data[1:]  # Rest is data
+
+        # 4. Create hash of file_key for consistent source_uuid prefix
+        file_hash = hashlib.md5(file_key.encode()).hexdigest()[:8]
+
+        # 5. Convert to queued_runs
+        queued_runs = []
+        for idx, row_values in enumerate(rows, 1):
+            # Pad row to match headers length
+            padded_row = row_values + [""] * (len(headers) - len(row_values))
+
+            # Create context variables dict
+            context_vars = dict(zip(headers, padded_row))
+
+            # Skip if no phone number
+            if not context_vars.get("phone_number"):
+                logger.debug(f"Skipping row {idx}: no phone_number")
+                continue
+
+            # Generate unique source UUID: csv_{hash(source_id)}_row_{idx}
+            source_uuid = f"csv_{file_hash}_row_{idx}"
+
+            queued_runs.append(
+                {
+                    "campaign_id": campaign_id,
+                    "source_uuid": source_uuid,
+                    "context_variables": context_vars,
+                    "state": "queued",
+                }
+            )
+
+        # 6. Bulk insert
+        if queued_runs:
+            await db_client.bulk_create_queued_runs(queued_runs)
+            logger.info(
+                f"Created {len(queued_runs)} queued runs for campaign {campaign_id}"
+            )
+
+        # 7. Update campaign total_rows
+        await db_client.update_campaign(
+            campaign_id=campaign_id,
+            total_rows=len(queued_runs),
+            source_sync_status="completed",
+        )
+
+        return len(queued_runs)
+
+    def _parse_csv(self, csv_content: str) -> List[List[str]]:
+        """Parse CSV content into rows"""
+        try:
+            csv_file = StringIO(csv_content)
+            reader = csv.reader(csv_file)
+            return list(reader)
+        except Exception as e:
+            logger.error(f"Failed to parse CSV: {e}")
+            raise ValueError(f"Invalid CSV format: {str(e)}")
+
+    async def validate_source_schema(self, source_config: Dict[str, Any]) -> bool:
+        """Validate that required columns exist in CSV"""
+        required_columns = ["phone_number", "first_name", "last_name"]
+
+        file_key = source_config.get("source_id")
+        if not file_key:
+            return False
+
+        # Get download URL using internal endpoint
+        signed_url = await storage_fs.aget_signed_url(
+            file_key, expiration=3600, use_internal_endpoint=True
+        )
+        if not signed_url:
+            return False
+
+        # Download just enough to get headers
+        async with httpx.AsyncClient() as client:
+            try:
+                response = await client.get(signed_url)
+                response.raise_for_status()
+
+                # Get just the first line for headers
+                first_line = response.text.split("\n")[0]
+                csv_file = StringIO(first_line)
+                reader = csv.reader(csv_file)
+                headers = next(reader, [])
+
+                return all(col in headers for col in required_columns)
+            except Exception as e:
+                logger.error(f"Failed to validate CSV schema: {e}")
+                return False

api/services/filesystem/base.py

@@ -33,13 +33,19 @@ class BaseFileSystem(ABC):
 
     @abstractmethod
     async def aget_signed_url(
-        self, file_path: str, expiration: int = 3600
+        self,
+        file_path: str,
+        expiration: int = 3600,
+        force_inline: bool = False,
+        use_internal_endpoint: bool = False,
     ) -> Optional[str]:
         """Generate a signed URL for temporary access to a file.
 
         Args:
             file_path: Path to the file
             expiration: URL expiration time in seconds (default: 1 hour)
+            force_inline: Force inline display (browser preview vs download)
+            use_internal_endpoint: Use internal endpoint (for container-to-container access)
 
         Returns:
             Optional[str]: Signed URL if successful, None otherwise
@@ -58,3 +64,24 @@ class BaseFileSystem(ABC):
             Contains: size, created_at, modified_at, etag, etc.
         """
         pass
+
+    @abstractmethod
+    async def aget_presigned_put_url(
+        self,
+        file_path: str,
+        expiration: int = 900,
+        content_type: str = "text/csv",
+        max_size: int = 10_485_760,
+    ) -> Optional[str]:
+        """Generate a presigned PUT URL for direct file upload.
+
+        Args:
+            file_path: Path where the file should be uploaded
+            expiration: URL expiration time in seconds (default: 15 minutes)
+            content_type: MIME type of the file (default: text/csv)
+            max_size: Maximum file size in bytes (default: 10MB)
+
+        Returns:
+            Optional[str]: Presigned PUT URL if successful, None otherwise
+        """
+        pass

api/services/filesystem/minio.py

@@ -48,17 +48,28 @@ class MinioFileSystem(BaseFileSystem):
             if not self.client.bucket_exists(self.bucket_name):
                 self.client.make_bucket(self.bucket_name)
 
-            # Set anonymous download policy for local development
-            # This allows unsigned URLs to work
+            # Set public read/write policy for local development
+            # This allows:
+            # 1. Anonymous downloads (s3:GetObject)
+            # 2. Anonymous uploads (s3:PutObject) - bypasses presigned URL signature issues
+            # 3. List bucket contents (s3:ListBucket) for debugging
+            # Note: This is set on every initialization to ensure policy is correct
+            # WARNING: Only use in local development, not production!
             policy = {
                 "Version": "2012-10-17",
                 "Statement": [
                     {
                         "Effect": "Allow",
                         "Principal": {"AWS": "*"},
-                        "Action": ["s3:GetObject"],
+                        "Action": ["s3:GetObject", "s3:PutObject", "s3:DeleteObject"],
                         "Resource": [f"arn:aws:s3:::{self.bucket_name}/*"],
-                    }
+                    },
+                    {
+                        "Effect": "Allow",
+                        "Principal": {"AWS": "*"},
+                        "Action": ["s3:ListBucket"],
+                        "Resource": [f"arn:aws:s3:::{self.bucket_name}"],
+                    },
                 ],
             }
 
@@ -97,14 +108,19 @@ class MinioFileSystem(BaseFileSystem):
             return False
 
     async def aget_signed_url(
-        self, file_path: str, expiration: int = 3600, force_inline: bool = False
+        self,
+        file_path: str,
+        expiration: int = 3600,
+        force_inline: bool = False,
+        use_internal_endpoint: bool = False,
     ) -> Optional[str]:
         try:
             # For MinIO in local development, return unsigned URLs
             # This avoids signature mismatch issues when endpoint differs
             # MinIO must be configured to allow anonymous read access
             protocol = "https" if self.secure else "http"
-            url = f"{protocol}://{self.public_endpoint}/{self.bucket_name}/{file_path}"
+            endpoint = self.endpoint if use_internal_endpoint else self.public_endpoint
+            url = f"{protocol}://{endpoint}/{self.bucket_name}/{file_path}"
             return url
         except Exception as e:
             logger.error(f"Error generating MinIO URL: {e}")
@@ -128,3 +144,29 @@ class MinioFileSystem(BaseFileSystem):
             }
         except S3Error:
             return None
+
+    async def aget_presigned_put_url(
+        self,
+        file_path: str,
+        expiration: int = 900,
+        content_type: str = "text/csv",
+        max_size: int = 10_485_760,
+    ) -> Optional[str]:
+        """Generate an unsigned URL for direct file upload.
+
+        For local MinIO development with anonymous upload enabled, we return
+        a simple unsigned URL instead of a presigned URL. This avoids signature
+        mismatch issues when the internal endpoint (minio:9000) differs from
+        the public endpoint (localhost:9000).
+
+        The bucket policy allows anonymous s3:PutObject, so no signature is needed.
+        """
+        try:
+            # Return unsigned URL for anonymous upload
+            protocol = "https" if self.secure else "http"
+            url = f"{protocol}://{self.public_endpoint}/{self.bucket_name}/{file_path}"
+            logger.debug(f"Generated unsigned upload URL: {url}")
+            return url
+        except Exception as e:
+            logger.error(f"Error generating MinIO upload URL: {e}")
+            return None

api/services/filesystem/s3.py

@@ -45,7 +45,11 @@ class S3FileSystem(BaseFileSystem):
             return False
 
     async def aget_signed_url(
-        self, file_path: str, expiration: int = 3600, force_inline: bool = False
+        self,
+        file_path: str,
+        expiration: int = 3600,
+        force_inline: bool = False,
+        use_internal_endpoint: bool = False,
     ) -> Optional[str]:
         """Generate a presigned GET url for the given object.
 
@@ -97,3 +101,28 @@ class S3FileSystem(BaseFileSystem):
                 }
         except ClientError:
             return None
+
+    async def aget_presigned_put_url(
+        self,
+        file_path: str,
+        expiration: int = 900,
+        content_type: str = "text/csv",
+        max_size: int = 10_485_760,
+    ) -> Optional[str]:
+        """Generate a presigned PUT URL for direct file upload."""
+        try:
+            async with self.session.client(
+                "s3", region_name=self.region_name
+            ) as s3_client:
+                url = await s3_client.generate_presigned_url(
+                    "put_object",
+                    Params={
+                        "Bucket": self.bucket_name,
+                        "Key": file_path,
+                        "ContentType": content_type,
+                    },
+                    ExpiresIn=expiration,
+                )
+            return url
+        except ClientError:
+            return None

api/services/telephony/ari_client_manager.py

@@ -426,16 +426,7 @@ async def setup_ari_client_supervisor(
 
     This is a drop-in replacement for the asyncari-based function.
     Uses AsyncARIClient instead of asyncari.
-
-    If the *ENABLE_ARI_STASIS* environment variable is not set to ``"true"``
-    (case-insensitive) the function returns ``None`` and no supervisor is
-    launched.
     """
-
-    if os.getenv("ENABLE_ARI_STASIS", "false").lower() != "true":
-        logger.info("ARI Stasis integration disabled via environment variable")
-        return None
-
     logger.info("Starting ARI Client Supervisor with AsyncARIClient")
 
     supervisor = _ARIClientManagerSupervisor(on_channel_start, on_channel_end)

api/services/telephony/ari_manager.py

@@ -16,7 +16,7 @@ import signal
 import time
 from typing import Dict, Optional
 
-from api.constants import REDIS_URL
+from api.constants import ENABLE_ARI_STASIS, REDIS_URL
 
 # --- Add logging setup before importing loguru ---
 from api.logging_config import setup_logging
@@ -32,7 +32,7 @@ from api.services.telephony.stasis_event_protocol import (
     parse_command,
 )
 
-logging_queue_listener = setup_logging()
+setup_logging()
 
 import redis.asyncio as aioredis
 import redis.exceptions
@@ -601,9 +601,13 @@ class ARIManager:
 
     async def run(self):
         """Main run loop for ARI Manager."""
-        self._running = True
+        if not ENABLE_ARI_STASIS:
+            logger.info("ARI Stasis integration disabled via environment variable")
+            return
 
         # Setup ARI connection with supervisor
+        self._running = True
+
         try:
             self._ari_client_supervisor = await setup_ari_client_supervisor(
                 self.on_channel_start, self.on_channel_end
@@ -737,9 +741,6 @@ async def main():
                     pass
     finally:
         await redis.aclose()
-        # --- Ensure Axiom logging listener is stopped gracefully ---
-        if logging_queue_listener is not None:
-            logging_queue_listener.stop()
 
 
 if __name__ == "__main__":

api/tasks/arq.py

@@ -9,7 +9,7 @@ from api.constants import REDIS_URL
 from api.logging_config import setup_logging
 from api.tasks.function_names import FunctionNames
 
-logging_queue_listener = setup_logging()
+setup_logging()
 
 # Now import ARQ and task dependencies
 from arq import create_pool

api/tasks/campaign_tasks.py

@@ -10,7 +10,7 @@ from api.services.campaign.campaign_event_protocol import BatchFailedEvent
 from api.services.campaign.campaign_event_publisher import (
     get_campaign_event_publisher,
 )
-from api.services.campaign.source_sync import get_sync_service
+from api.services.campaign.source_sync_factory import get_sync_service
 
 
 async def sync_campaign_source(ctx: Dict, campaign_id: int) -> None:

docker-compose.yaml

@@ -45,6 +45,7 @@ services:
     environment:
       MINIO_ROOT_USER: minioadmin
       MINIO_ROOT_PASSWORD: minioadmin
+      MINIO_API_CORS_ALLOW_ORIGIN: "*"
     ports:
       - "127.0.0.1:9000:9000" # Bind to localhost explicitly
       - "127.0.0.1:9001:9001"
@@ -90,8 +91,8 @@ services:
       # LANGFUSE_HOST: "https://langfuse.dograh.com"
 
       # Sentry
-      ENABLE_SETRY: "false"
-      SENTRY_DSN: ""
+      ENABLE_TELEMETRY: "${ENABLE_TELEMETRY:-true}"
+      SENTRY_DSN: "https://3acdb63d5f1f70430953353b82de61e0@o4509486225096704.ingest.us.sentry.io/4510152922693632"
 
     ports:
       - "8000:8000"
@@ -104,12 +105,6 @@ services:
         condition: service_healthy
       cloudflared:
         condition: service_started
-    command: >
-      bash -c "
-        cd /app/api &&
-        alembic upgrade head &&
-        uvicorn api.app:app --host 0.0.0.0 --port 8000
-      "
     healthcheck:
       test:
         [
@@ -123,49 +118,6 @@ services:
     networks:
       - app-network
 
-  arq-worker:
-    image: ${REGISTRY:-dograhai}/dograh-api:latest
-    volumes:
-      - shared-tmp:/tmp
-    environment:
-      # Core application config
-      ENVIRONMENT: "local"
-      LOG_LEVEL: "INFO"
-
-      # Database configuration (using containerized postgres)
-      DATABASE_URL: "postgresql+asyncpg://postgres:postgres@postgres:5432/postgres"
-
-      # Redis configuration (using containerized redis)
-      REDIS_URL: "redis://:redissecret@redis:6379"
-
-      # Storage configuration - using local MinIO
-      ENABLE_AWS_S3: "false"
-
-      # MinIO
-      MINIO_ENDPOINT: "minio:9000"
-      MINIO_ACCESS_KEY: "minioadmin"
-      MINIO_SECRET_KEY: "minioadmin"
-      MINIO_BUCKET: "voice-audio"
-      MINIO_SECURE: "false"
-
-      # Sentry
-      ENABLE_SETRY: "false"
-      SENTRY_DSN: ""
-    command: >
-      bash -c "
-        cd /app/api &&
-        python -m arq api.tasks.arq.WorkerSettings
-      "
-    depends_on:
-      postgres:
-        condition: service_healthy
-      redis:
-        condition: service_healthy
-      minio:
-        condition: service_healthy
-    networks:
-      - app-network
-
   ui:
     image: ${REGISTRY:-dograhai}/dograh-ui:latest
     environment:

scripts/rolling_update_uvicorn.sh

@@ -4,18 +4,22 @@
 set -e  # Exit on error
 
 ### CONFIGURATION #############################################################
-ENV_FILE="api/.env"
-RUN_DIR="run"
-BASE_LOG_DIR="/home/ubuntu/dograh/logs"  # Base logs directory (same as start_services.sh)
+
+# Determine BASE_DIR as parent of the scripts directory
+BASE_DIR="$(cd "$(dirname "$(dirname "${BASH_SOURCE[0]}")")" && pwd)"
+
+ENV_FILE="$BASE_DIR/api/.env"
+RUN_DIR="$BASE_DIR/run"
+BASE_LOG_DIR="$BASE_DIR/logs"  # Base logs directory (same as start_services.sh)
 LATEST_LINK="$BASE_LOG_DIR/latest"        # Symlink to latest logs (same as start_services.sh)
-VENV_PATH="/home/ubuntu/dograh/venv"
+VENV_PATH="$BASE_DIR/venv"
 HEALTH_CHECK_ENDPOINT="/api/v1/health"  # Adjust as needed
 MAX_WAIT_SECONDS=310  # Max wait for graceful shutdown (5 minutes + 10 seconds grace)
 
 # Load environment
 set -a && . "$ENV_FILE" && set +a
 
-cd /home/ubuntu/dograh/app
+cd "$BASE_DIR"
 
 ### FUNCTIONS ##################################################################
 
@@ -166,10 +170,13 @@ start_new_uvicorn_workers() {
     
     log_info "Starting uvicorn with $FASTAPI_WORKERS workers on port $new_port"
     log_info "Logs: $LOG_FILE_PATH"
-    
-    # Start in new process group with setsid (same as start_services.sh)
-    # Each service gets its own LOG_FILE_PATH environment variable
-    setsid nohup bash -c "LOG_FILE_PATH='$LOG_FILE_PATH' uvicorn api.app:app --host 0.0.0.0 --port $new_port --workers $FASTAPI_WORKERS" >/dev/null 2>&1 &
+
+    # Start in background (same pattern as start_services.sh)
+    (
+        cd "$BASE_DIR"
+        export LOG_FILE_PATH="$log_dir/uvicorn-rollover-${timestamp}-${script_pid}.log"
+        exec uvicorn api.app:app --host 0.0.0.0 --port $new_port --workers $FASTAPI_WORKERS >>"$LOG_FILE_PATH" 2>&1
+    ) &
 
     local new_pid=$!
     echo "$new_pid" > "$RUN_DIR/uvicorn_new.pid"

scripts/start_services.sh

@@ -1,70 +1,101 @@
 #!/usr/bin/env bash
-# start_services.sh
-
 set -e  # Exit on error
 
-### CONFIGURATION #############################################################
-ENV_FILE="api/.env"
-RUN_DIR="run"                 # where we keep *.pid
-BASE_LOG_DIR="/home/ubuntu/dograh/logs"           # base logs directory
+###############################################################################
+### CONFIGURATION
+###############################################################################
+
+# Determine BASE_DIR as parent of the scripts directory
+BASE_DIR="$(cd "$(dirname "$(dirname "${BASH_SOURCE[0]}")")" && pwd)"
+
+ENV_FILE="$BASE_DIR/api/.env"
+RUN_DIR="$BASE_DIR/run"                 # Where we keep *.pid
+BASE_LOG_DIR="$BASE_DIR/logs"           # Base logs directory
+
 TIMESTAMP=$(date +"%Y%m%d_%H%M%S")
-LOG_DIR="$BASE_LOG_DIR/$TIMESTAMP"  # timestamped log directory
-LATEST_LINK="$BASE_LOG_DIR/latest"  # symlink to latest logs
-VENV_PATH="/home/ubuntu/dograh/venv"
+LOG_DIR="$BASE_LOG_DIR/$TIMESTAMP"      # Timestamped log directory
+LATEST_LINK="$BASE_LOG_DIR/latest"      # Symlink to latest logs
+VENV_PATH="$BASE_DIR/venv"
+
 ARQ_WORKERS=${ARQ_WORKERS:-1}
 
 # Log startup
-echo "Starting Dograh Services at $(date)"
+cd "$BASE_DIR"
+echo "Starting Dograh Services at $(date) in BASE_DIR: ${BASE_DIR}"
 
-### 1) Load environment vars so that configurations like FASTAPI_WORKERS are loaded
-set -a && . "$ENV_FILE" && set +a
+###############################################################################
+### 1) Load environment variables
+###############################################################################
 
-cd /home/ubuntu/dograh/app
-
-if [[ -z "${FASTAPI_PORT:-}" ]]; then
-  echo "Error: FASTAPI_PORT environment variable is not set."
-  exit 1
+# Load environment from a file if it exists
+if [[ -f "$ENV_FILE" ]]; then
+  set -a && . "$ENV_FILE" && set +a
 fi
 
-if [[ -z "${FASTAPI_WORKERS:-}" ]]; then
-  echo "Error: FASTAPI_WORKERS environment variable is not set."
-  exit 1
-fi
+FASTAPI_PORT=${FASTAPI_PORT:-8000}
+FASTAPI_WORKERS=${FASTAPI_WORKERS:-1}
 
-# map "service name" → "command to run"
-declare -A SERVICES=(
-  [ari_manager]="python -m api.services.telephony.ari_manager"
-  [campaign_orchestrator]="python -m api.services.campaign.campaign_orchestrator"
-  [uvicorn]="uvicorn api.app:app --host 0.0.0.0 --port $FASTAPI_PORT --workers $FASTAPI_WORKERS"
+###############################################################################
+### 2) Define services
+###############################################################################
+
+# Map "service name" → "command to run"
+# Using arrays for bash 3.2 compatibility
+SERVICE_NAMES=(
+  "ari_manager"
+  "campaign_orchestrator"
+  "uvicorn"
 )
 
-# Add ARQ workers dynamically based on ARQ_WORKERS environment variable
+SERVICE_COMMANDS=(
+  "python -m api.services.telephony.ari_manager"
+  "python -m api.services.campaign.campaign_orchestrator"
+  "uvicorn api.app:app --host 0.0.0.0 --port $FASTAPI_PORT --workers $FASTAPI_WORKERS"
+)
+
+# Add ARQ workers dynamically
 for ((i=1; i<=ARQ_WORKERS; i++)); do
-  SERVICES[arq$i]="python -m arq api.tasks.arq.WorkerSettings --custom-log-dict api.tasks.arq.LOG_CONFIG"
+  SERVICE_NAMES+=("arq$i")
+  SERVICE_COMMANDS+=("python -m arq api.tasks.arq.WorkerSettings --custom-log-dict api.tasks.arq.LOG_CONFIG")
 done
 
-### 2) Activate virtual environment #########################################
-source ${VENV_PATH}/bin/activate
+###############################################################################
+### 3) Activate virtual environment
+###############################################################################
+
+if [[ -d "$VENV_PATH" && -f "$VENV_PATH/bin/activate" ]]; then
+  source "$VENV_PATH/bin/activate"
+  echo "Virtual environment activated: $VENV_PATH"
+else
+  echo "Warning: Virtual environment not found at $VENV_PATH"
+  echo "Continuing without virtual environment activation..."
+fi
+
+###############################################################################
+### 4) Stop old services
+###############################################################################
 
-### 3) Stop old services (only via PID files) #################################
 mkdir -p "$RUN_DIR"
-for name in "${!SERVICES[@]}"; do
+for name in "${SERVICE_NAMES[@]}"; do
   pidfile="$RUN_DIR/$name.pid"
+
   if [[ -f $pidfile ]]; then
     oldpid=$(<"$pidfile")
-    if kill -0 "$oldpid"; then
+
+    if kill -0 "$oldpid" 2>/dev/null; then
       echo "Stopping $name (PID $oldpid and its process group)…"
+
       # Kill the entire process group (negative PID)
-      # First try SIGTERM
-      kill -TERM -"$oldpid" || kill -TERM "$oldpid" || true
+      kill -TERM -"$oldpid" 2>/dev/null || kill -TERM "$oldpid" 2>/dev/null || true
       sleep 4
-      # If still running, use SIGKILL
-      if kill -0 "$oldpid"; then
+
+      if kill -0 "$oldpid" 2>/dev/null; then
         echo "⚠️  $name did not exit cleanly, forcing stop..."
-        kill -KILL -"$oldpid" || kill -KILL "$oldpid" || true
+        kill -KILL -"$oldpid" 2>/dev/null || kill -KILL "$oldpid" 2>/dev/null || true
         sleep 1
       fi
     fi
+
     rm -f "$pidfile"
   else
     echo "No PID file for $name, skipping stop."
@@ -74,14 +105,19 @@ done
 # Clean up any port tracking files for uvicorn
 rm -f "$RUN_DIR/uvicorn.port" "$RUN_DIR/uvicorn_new.port" "$RUN_DIR/uvicorn_old.pid"
 
-### 4) Run migrations #########################################################
-alembic -c api/alembic.ini upgrade head
+###############################################################################
+### 5) Run migrations
+###############################################################################
 
-### 5) Prepare logs ###########################################################
-mkdir -p "$BASE_LOG_DIR"
-mkdir -p "$LOG_DIR"
+alembic -c "$BASE_DIR/api/alembic.ini" upgrade head
 
-# Remove old symlink if it exists and create new one
+###############################################################################
+### 6) Prepare logs
+###############################################################################
+
+mkdir -p "$BASE_LOG_DIR" "$LOG_DIR"
+
+# Remove old symlink and create a new one
 if [[ -L "$LATEST_LINK" ]]; then
   rm "$LATEST_LINK"
 fi
@@ -90,33 +126,37 @@ ln -s "$TIMESTAMP" "$LATEST_LINK"
 echo "Log directory: $LOG_DIR"
 echo "Latest symlink: $LATEST_LINK -> $TIMESTAMP"
 
-### 7) Start services #########################################################
-for name in "${!SERVICES[@]}"; do
-  cmd=${SERVICES[$name]}
+###############################################################################
+### 7) Start services
+###############################################################################
+
+for i in "${!SERVICE_NAMES[@]}"; do
+  name="${SERVICE_NAMES[$i]}"
+  cmd="${SERVICE_COMMANDS[$i]}"
   echo "→ Starting $name"
-  
-  # Export LOG_FILE_PATH for this specific service
-  export LOG_FILE_PATH="$LOG_DIR/$name.log"
-  
-  # Start in new process group with setsid
-  # Each service gets its own LOG_FILE_PATH environment variable
-  setsid nohup bash -c "LOG_FILE_PATH='$LOG_DIR/$name.log' $cmd" >/dev/null 2>&1 &
-  
-  # Get the PID of the setsid process
+
+  (
+    cd "$BASE_DIR"
+    export LOG_FILE_PATH="$LOG_DIR/$name.log"
+    exec $cmd >>"$LOG_DIR/$name.log" 2>&1
+  ) &
+
   pid=$!
   echo $pid >"$RUN_DIR/$name.pid"
-  
-  # For uvicorn, also save the port for rolling updates
+  echo "  Started with PID $pid"
+
   if [[ "$name" == "uvicorn" ]]; then
     echo "$FASTAPI_PORT" >"$RUN_DIR/uvicorn.port"
   fi
 done
-disown -a
 
-### 8) Summary #################################################################
+###############################################################################
+### 8) Summary
+###############################################################################
+
 echo
 echo "──────────────────────────────────────────────────"
-for name in "${!SERVICES[@]}"; do
+for name in "${SERVICE_NAMES[@]}"; do
   pid=$(<"$RUN_DIR/$name.pid")
   echo "✓ $name (PID $pid) → $LOG_DIR/$name.log"
 done

ui/src/app/campaigns/CsvUploadSelector.tsx

@@ -0,0 +1,139 @@
+'use client';
+
+import { useRef, useState } from 'react';
+import { toast } from 'sonner';
+
+import { Button } from '@/components/ui/button';
+import { Label } from '@/components/ui/label';
+import logger from '@/lib/logger';
+
+interface CsvUploadSelectorProps {
+  accessToken: string;
+  onFileUploaded: (fileKey: string, fileName: string) => void;
+  selectedFileName?: string;
+}
+
+interface PresignedUploadUrlResponse {
+  upload_url: string;
+  file_key: string;
+  expires_in: number;
+}
+
+const MAX_FILE_SIZE = 10 * 1024 * 1024; // 10MB
+
+export default function CsvUploadSelector({ accessToken, onFileUploaded, selectedFileName }: CsvUploadSelectorProps) {
+  const [uploading, setUploading] = useState(false);
+  const [uploadProgress, setUploadProgress] = useState(0);
+  const fileInputRef = useRef<HTMLInputElement>(null);
+
+  const handleFileSelect = async (event: React.ChangeEvent<HTMLInputElement>) => {
+    const file = event.target.files?.[0];
+    if (!file) return;
+
+    // Validate file type
+    if (!file.name.endsWith('.csv')) {
+      toast.error('Please select a CSV file');
+      return;
+    }
+
+    // Validate file size
+    if (file.size > MAX_FILE_SIZE) {
+      toast.error('File size must be less than 10MB');
+      return;
+    }
+
+    setUploading(true);
+    setUploadProgress(0);
+
+    try {
+      // Step 1: Request presigned upload URL
+      logger.info('Requesting presigned upload URL for:', file.name);
+      const presignedResponse = await fetch('/api/v1/s3/presigned-upload-url', {
+        method: 'POST',
+        headers: {
+          'Authorization': `Bearer ${accessToken}`,
+          'Content-Type': 'application/json',
+        },
+        body: JSON.stringify({
+          file_name: file.name,
+          file_size: file.size,
+          content_type: 'text/csv',
+        }),
+      });
+
+      if (!presignedResponse.ok) {
+        const error = await presignedResponse.json();
+        throw new Error(error.detail || 'Failed to get upload URL');
+      }
+
+      const presignedData: PresignedUploadUrlResponse = await presignedResponse.json();
+      logger.info('Received presigned URL, uploading file...');
+
+      // Step 2: Upload file directly to S3/MinIO
+      const uploadResponse = await fetch(presignedData.upload_url, {
+        method: 'PUT',
+        body: file,
+        headers: {
+          'Content-Type': 'text/csv',
+        },
+      });
+
+      if (!uploadResponse.ok) {
+        throw new Error('Failed to upload file to storage');
+      }
+
+      setUploadProgress(100);
+      logger.info('File uploaded successfully, file_key:', presignedData.file_key);
+
+      // Step 3: Notify parent with file_key
+      onFileUploaded(presignedData.file_key, file.name);
+      toast.success(`File uploaded: ${file.name}`);
+    } catch (error) {
+      logger.error('Error uploading CSV:', error);
+      toast.error(error instanceof Error ? error.message : 'Failed to upload CSV file');
+    } finally {
+      setUploading(false);
+      setUploadProgress(0);
+      // Reset file input
+      if (fileInputRef.current) {
+        fileInputRef.current.value = '';
+      }
+    }
+  };
+
+  const handleButtonClick = () => {
+    fileInputRef.current?.click();
+  };
+
+  return (
+    <div className="space-y-2">
+      <Label>CSV File</Label>
+      <div className="flex items-center gap-4">
+        <input
+          ref={fileInputRef}
+          type="file"
+          accept=".csv"
+          onChange={handleFileSelect}
+          className="hidden"
+        />
+        <Button
+          type="button"
+          variant="outline"
+          onClick={handleButtonClick}
+          disabled={uploading}
+        >
+          {uploading ? `Uploading... ${uploadProgress}%` : 'Upload CSV File'}
+        </Button>
+        {selectedFileName && !uploading && (
+          <div className="flex-1 text-sm">
+            <span className="text-gray-600">Selected: </span>
+            <span className="text-blue-600">{selectedFileName}</span>
+          </div>
+        )}
+      </div>
+      <p className="text-sm text-gray-500">
+        Upload a CSV file with contact data. Must include phone_number, first_name, and last_name columns. Max 10MB.
+      </p>
+    </div>
+  );
+}

ui/src/app/campaigns/[campaignId]/page.tsx

@@ -8,6 +8,7 @@ import { toast } from 'sonner';
 import {
     getCampaignApiV1CampaignCampaignIdGet,
     getCampaignRunsApiV1CampaignCampaignIdRunsGet,
+    getCampaignSourceDownloadUrlApiV1CampaignCampaignIdSourceDownloadUrlGet,
     pauseCampaignApiV1CampaignCampaignIdPausePost,
     resumeCampaignApiV1CampaignCampaignIdResumePost,
     startCampaignApiV1CampaignCampaignIdStartPost} from '@/client/sdk.gen';
@@ -125,6 +126,33 @@ export default function CampaignDetailPage() {
         }
     };
 
+    // Handle CSV download
+    const handleDownloadCsv = async () => {
+        if (!user || !campaign || campaign.source_type !== 'csv') return;
+
+        try {
+            const accessToken = await getAccessToken();
+            const response = await getCampaignSourceDownloadUrlApiV1CampaignCampaignIdSourceDownloadUrlGet({
+                path: {
+                    campaign_id: campaignId,
+                },
+                headers: {
+                    'Authorization': `Bearer ${accessToken}`,
+                }
+            });
+
+            if (response.data?.download_url) {
+                // Open download URL in new tab
+                window.open(response.data.download_url, '_blank');
+            } else {
+                toast.error('Failed to get download URL');
+            }
+        } catch (error) {
+            console.error('Failed to download CSV:', error);
+            toast.error('Failed to download CSV file');
+        }
+    };
+
     // Handle start campaign
     const handleStart = async () => {
         if (!user) return;
@@ -354,16 +382,27 @@ export default function CampaignDetailPage() {
                                 <dd className="mt-1 capitalize">{campaign.source_type.replace('-', ' ')}</dd>
                             </div>
                             <div>
-                                <dt className="text-sm font-medium text-gray-500">Source Sheet</dt>
+                                <dt className="text-sm font-medium text-gray-500">
+                                    {campaign.source_type === 'csv' ? 'Source File' : 'Source Sheet'}
+                                </dt>
                                 <dd className="mt-1">
-                                    <a
-                                        href={campaign.source_id}
-                                        target="_blank"
-                                        rel="noopener noreferrer"
-                                        className="text-blue-600 hover:text-blue-800 hover:underline text-sm break-all"
-                                    >
-                                        {campaign.source_id}
-                                    </a>
+                                    {campaign.source_type === 'csv' ? (
+                                        <button
+                                            onClick={handleDownloadCsv}
+                                            className="text-blue-600 hover:text-blue-800 hover:underline text-sm break-all"
+                                        >
+                                            {campaign.source_id.split('/').pop()}
+                                        </button>
+                                    ) : (
+                                        <a
+                                            href={campaign.source_id}
+                                            target="_blank"
+                                            rel="noopener noreferrer"
+                                            className="text-blue-600 hover:text-blue-800 hover:underline text-sm break-all"
+                                        >
+                                            {campaign.source_id}
+                                        </a>
+                                    )}
                                 </dd>
                             </div>
                             <div>

ui/src/app/campaigns/new/page.tsx

@@ -20,6 +20,7 @@ import {
 } from '@/components/ui/select';
 import { useAuth } from '@/lib/auth';
 
+import CsvUploadSelector from '../CsvUploadSelector';
 import GoogleSheetSelector from '../GoogleSheetSelector';
 
 export default function NewCampaignPage() {
@@ -29,7 +30,9 @@ export default function NewCampaignPage() {
     // Form state
     const [campaignName, setCampaignName] = useState('');
     const [selectedWorkflowId, setSelectedWorkflowId] = useState<string>('');
-    const [selectedSheetUrl, setSelectedSheetUrl] = useState('');
+    const [sourceType, setSourceType] = useState<'google-sheet' | 'csv'>('csv');
+    const [sourceId, setSourceId] = useState('');
+    const [selectedFileName, setSelectedFileName] = useState('');
     const [isSubmitting, setIsSubmitting] = useState(false);
     const [userAccessToken, setUserAccessToken] = useState<string>('');
 
@@ -78,7 +81,7 @@ export default function NewCampaignPage() {
     const handleSubmit = async (e: React.FormEvent) => {
         e.preventDefault();
 
-        if (!campaignName || !selectedWorkflowId || !selectedSheetUrl) {
+        if (!campaignName || !selectedWorkflowId || !sourceId) {
             toast.error('Please fill in all fields');
             return;
         }
@@ -91,7 +94,8 @@ export default function NewCampaignPage() {
                 body: {
                     name: campaignName,
                     workflow_id: parseInt(selectedWorkflowId),
-                    source_id: selectedSheetUrl,
+                    source_type: sourceType,
+                    source_id: sourceId,
                 },
                 headers: {
                     'Authorization': `Bearer ${accessToken}`,
@@ -117,7 +121,13 @@ export default function NewCampaignPage() {
 
     // Handle sheet selection
     const handleSheetSelected = (sheetUrl: string) => {
-        setSelectedSheetUrl(sheetUrl);
+        setSourceId(sheetUrl);
+    };
+
+    // Handle CSV file upload
+    const handleFileUploaded = (fileKey: string, fileName: string) => {
+        setSourceId(fileKey);
+        setSelectedFileName(fileName);
     };
 
     return (
@@ -191,20 +201,52 @@ export default function NewCampaignPage() {
                                     </SelectContent>
                                 </Select>
                                 <p className="text-sm text-gray-500">
-                                    Select the workflow to execute for each row in the spreadsheet
+                                    Select the workflow to execute for each row in the data source
                                 </p>
                             </div>
 
-                            <GoogleSheetSelector
-                                accessToken={userAccessToken}
-                                onSheetSelected={handleSheetSelected}
-                                selectedSheetUrl={selectedSheetUrl}
-                            />
+                            <div className="space-y-2">
+                                <Label htmlFor="source-type">Data Source Type</Label>
+                                <Select
+                                    value={sourceType}
+                                    onValueChange={(value) => {
+                                        setSourceType(value as 'google-sheet' | 'csv');
+                                        setSourceId('');
+                                        setSelectedFileName('');
+                                    }}
+                                    required
+                                >
+                                    <SelectTrigger id="source-type">
+                                        <SelectValue placeholder="Select source type" />
+                                    </SelectTrigger>
+                                    <SelectContent>
+                                        <SelectItem value="google-sheet">Google Sheet</SelectItem>
+                                        <SelectItem value="csv">CSV File</SelectItem>
+                                    </SelectContent>
+                                </Select>
+                                <p className="text-sm text-gray-500">
+                                    Choose where your contact data is stored
+                                </p>
+                            </div>
+
+                            {sourceType === 'google-sheet' ? (
+                                <GoogleSheetSelector
+                                    accessToken={userAccessToken}
+                                    onSheetSelected={handleSheetSelected}
+                                    selectedSheetUrl={sourceId}
+                                />
+                            ) : (
+                                <CsvUploadSelector
+                                    accessToken={userAccessToken}
+                                    onFileUploaded={handleFileUploaded}
+                                    selectedFileName={selectedFileName}
+                                />
+                            )}
 
                             <div className="flex gap-4 pt-4">
                                 <Button
                                     type="submit"
-                                    disabled={isSubmitting || !campaignName || !selectedWorkflowId || !selectedSheetUrl}
+                                    disabled={isSubmitting || !campaignName || !selectedWorkflowId || !sourceId}
                                 >
                                     {isSubmitting ? 'Creating...' : 'Create Campaign'}
                                 </Button>

ui/src/client/types.gen.ts

@@ -72,6 +72,11 @@ export type CampaignResponse = {
     completed_at: string | null;
 };
 
+export type CampaignSourceDownloadResponse = {
+    download_url: string;
+    expires_in: number;
+};
+
 export type CampaignsResponse = {
     campaigns: Array<CampaignResponse>;
 };
@@ -91,6 +96,7 @@ export type CreateApiKeyResponse = {
 export type CreateCampaignRequest = {
     name: string;
     workflow_id: number;
+    source_type: string;
     source_id: string;
 };
 
@@ -287,6 +293,27 @@ export type LoadTestStatsResponse = {
     }>;
 };
 
+export type PresignedUploadUrlRequest = {
+    /**
+     * CSV filename
+     */
+    file_name: string;
+    /**
+     * File size in bytes (max 10MB)
+     */
+    file_size: number;
+    /**
+     * File content type
+     */
+    content_type?: string;
+};
+
+export type PresignedUploadUrlResponse = {
+    upload_url: string;
+    file_key: string;
+    expires_in: number;
+};
+
 export type RtcOfferRequest = {
     pc_id: string | null;
     sdp: string;
@@ -1758,6 +1785,40 @@ export type GetCampaignProgressApiV1CampaignCampaignIdProgressGetResponses = {
 
 export type GetCampaignProgressApiV1CampaignCampaignIdProgressGetResponse = GetCampaignProgressApiV1CampaignCampaignIdProgressGetResponses[keyof GetCampaignProgressApiV1CampaignCampaignIdProgressGetResponses];
 
+export type GetCampaignSourceDownloadUrlApiV1CampaignCampaignIdSourceDownloadUrlGetData = {
+    body?: never;
+    headers?: {
+        authorization?: string | null;
+    };
+    path: {
+        campaign_id: number;
+    };
+    query?: never;
+    url: '/api/v1/campaign/{campaign_id}/source-download-url';
+};
+
+export type GetCampaignSourceDownloadUrlApiV1CampaignCampaignIdSourceDownloadUrlGetErrors = {
+    /**
+     * Not found
+     */
+    404: unknown;
+    /**
+     * Validation Error
+     */
+    422: HttpValidationError;
+};
+
+export type GetCampaignSourceDownloadUrlApiV1CampaignCampaignIdSourceDownloadUrlGetError = GetCampaignSourceDownloadUrlApiV1CampaignCampaignIdSourceDownloadUrlGetErrors[keyof GetCampaignSourceDownloadUrlApiV1CampaignCampaignIdSourceDownloadUrlGetErrors];
+
+export type GetCampaignSourceDownloadUrlApiV1CampaignCampaignIdSourceDownloadUrlGetResponses = {
+    /**
+     * Successful Response
+     */
+    200: CampaignSourceDownloadResponse;
+};
+
+export type GetCampaignSourceDownloadUrlApiV1CampaignCampaignIdSourceDownloadUrlGetResponse = GetCampaignSourceDownloadUrlApiV1CampaignCampaignIdSourceDownloadUrlGetResponses[keyof GetCampaignSourceDownloadUrlApiV1CampaignCampaignIdSourceDownloadUrlGetResponses];
+
 export type GetIntegrationsApiV1IntegrationGetData = {
     body?: never;
     headers?: {
@@ -2028,6 +2089,38 @@ export type GetFileMetadataApiV1S3FileMetadataGetResponses = {
 
 export type GetFileMetadataApiV1S3FileMetadataGetResponse = GetFileMetadataApiV1S3FileMetadataGetResponses[keyof GetFileMetadataApiV1S3FileMetadataGetResponses];
 
+export type GetPresignedUploadUrlApiV1S3PresignedUploadUrlPostData = {
+    body: PresignedUploadUrlRequest;
+    headers?: {
+        authorization?: string | null;
+    };
+    path?: never;
+    query?: never;
+    url: '/api/v1/s3/presigned-upload-url';
+};
+
+export type GetPresignedUploadUrlApiV1S3PresignedUploadUrlPostErrors = {
+    /**
+     * Not found
+     */
+    404: unknown;
+    /**
+     * Validation Error
+     */
+    422: HttpValidationError;
+};
+
+export type GetPresignedUploadUrlApiV1S3PresignedUploadUrlPostError = GetPresignedUploadUrlApiV1S3PresignedUploadUrlPostErrors[keyof GetPresignedUploadUrlApiV1S3PresignedUploadUrlPostErrors];
+
+export type GetPresignedUploadUrlApiV1S3PresignedUploadUrlPostResponses = {
+    /**
+     * Successful Response
+     */
+    200: PresignedUploadUrlResponse;
+};
+
+export type GetPresignedUploadUrlApiV1S3PresignedUploadUrlPostResponse = GetPresignedUploadUrlApiV1S3PresignedUploadUrlPostResponses[keyof GetPresignedUploadUrlApiV1S3PresignedUploadUrlPostResponses];
+
 export type GetServiceKeysApiV1UserServiceKeysGetData = {
     body?: never;
     headers?: {